Learning in public, still figuring it out
Why I'm writing down what I learn, including the parts I don't understand yet.
I’ll be honest about something: a lot of the time, I’m figuring this out as I go.
That’s part of why this log exists. I’d rather write down what I’m learning, including the messy parts, than pretend I have it all solved.
What I’m exploring right now
On my own time, I’ve been running AI on hardware I control instead of only using it through someone else’s service.
I’m not doing it because I think everyone should. I’m doing it because I want to understand what changes when you do. What gets easier, what gets harder, and what I didn’t expect.
So far the answer is: lots of reading, and lots of breaking things.
Why break things on purpose
The fastest way I learn is by getting stuck. Something doesn’t work, I dig in, and suddenly I understand a piece of it for real instead of just nodding along.
At work, I can’t break things. There are real businesses on the other end. So the side project is where I get to be clumsy.
Why write it down
A few reasons.
- It makes me finish the thought. If I can’t explain it in plain words, I don’t get it yet.
- It’s a record. Future me will forget why I made a choice. Now he can look it up.
- It might help someone. If you’re also figuring this out, you’re not the only one.
What I won’t do here
I won’t pretend to know more than I do. If something is a guess, I’ll say it’s a guess. If I got something wrong last month, I’ll say that too.
My tools aren’t allowed to make things up or puff them up, and if they’re unsure, they say so. It’s only fair to hold myself to it.
What’s next
More experiments, more notes, probably more broken things. I’m eager to learn, and this is where I’ll write it down.
If any of it turns out useful, great. If not, at least I’ll understand it a little better than I did yesterday.